{"id":100,"date":"2006-05-25T16:28:43","date_gmt":"2006-05-25T13:28:43","guid":{"rendered":"http:\/\/www.muratakyildiz.com\/wordpress\/?p=100"},"modified":"2014-09-11T18:46:44","modified_gmt":"2014-09-11T15:46:44","slug":"aratirma-modeller-ve-kullanilan-statstkler","status":"publish","type":"post","link":"https:\/\/www.istatistik.gen.tr\/?p=100","title":{"rendered":"ARA\u015eTIRMA MODELLER\u0130 VE KULLANILAN \u0130STAT\u0130ST\u0130KLER"},"content":{"rendered":"

ARA\u015eTIRMA MODELLER\u0130 VE KULLANILAN \u0130STAT\u0130ST\u0130KLER
\n<\/span><\/span><\/strong><\/p>\n

 <\/p>\n

 <\/p>\n

 <\/p>\n\n\n\n\n\n\n\n
ARA\u015fTIRMA MODELLER\u0131
\n<\/strong><\/td>\n<\/tr>\n
A) Kontrol Grubu ve \u00d6rneklemi Olmayan Modeller
\n<\/span><\/span><\/strong><\/td>\n		B) Kontrol Grubu Olan Modeller
\n<\/span><\/span><\/strong><\/td>\n		C) \u00d6rneklem Ve Kontrol Gruplu Modeller
\n<\/span><\/span><\/strong><\/td>\n<\/tr>\n
1-<\/span> Kontrols\u00fcz Son Test Modeli:
\n<\/span><\/span><\/td>\n		1- Kontroll\u00fc son test modeli:
\n<\/span><\/span><\/td>\n		1- \u00c7apraz Test Modeli:
\n<\/span><\/span><\/td>\n<\/tr>\n
2-Kontrols\u00fcz \u00d6n ve Son Test Modeli:
\n<\/span><\/span><\/td>\n		2- Kontroll\u00fc \u00f6n test ve son test modeli:
\n<\/span><\/span><\/td>\n		2- \u00d6rneklem ve kontroll\u00fc, \u00d6n ve Son test Modeli:
\n<\/span><\/span><\/td>\n<\/tr>\n
<\/td>\n<\/td>\n3- \u00d6rneklem ve kontroll\u00fc, Son test Modeli:
\n<\/span><\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n

<\/p>\n

K\u0131saltmalar:
\n<\/span><\/span><\/span><\/p>\n

G : Grup <\/span><\/p>\n

Gd<\/sub> : Deney Grubu:<\/span><\/p>\n

Gk<\/sub> : Kontrol Grubu:<\/span><\/p>\n

T : Test<\/span><\/p>\n

GR<\/sup> : Rassal \u00f6rnekleme teknikleri ile se\u00e7ilen grup<\/span><\/p>\n

T\u00f6<\/sub> : \u00d6n test:<\/span><\/p>\n

Ts : Son test:
\n<\/sub><\/span><\/span><\/p>\n

D: Deney<\/span><\/p>\n

 <\/p>\n

 <\/p>\n

A)<\/span> <\/strong>KONTROL GRUBU VE \u00d6RNEKLEM\u0130 OLMAYAN MODELLER
\n<\/span><\/strong><\/span><\/p>\n

 <\/p>\n

 <\/p>\n

Bu tip modellerde \u00f6rnekleme yoktur, deneye\/teste tabi tutulacak grup vard\u0131r.<\/span><\/p>\n\n\n\n
\n
\n

G D T
\n<\/span><\/span><\/strong><\/p>\n<\/div>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n


\n<\/strong>1-<\/span> <\/strong>Kontrols\u00fcz Son Test Modeli:
\n<\/span><\/strong><\/span><\/p>\n

 <\/p>\n

 <\/p>\n

Bir gruba denemesi yap\u0131lacak i\u015flem uygulan\u0131r. Test\/deney sonunda elde edilen sonu\u00e7lar\u0131n anlaml\u0131 olup olmad\u0131\u011f\u0131 \u201ctek gruplar i\u00e7in t testi\u201d ile s\u0131nan\u0131r. <\/span><\/p>\n

 <\/p>\n

 <\/p>\n


\n<\/strong><\/p>\n

 <\/p>\n

 <\/p>\n

 <\/p>\n\n\n\n
\n
\n

G T\u00f6<\/sub> D Ts
\n<\/sub><\/span><\/span><\/strong><\/p>\n<\/div>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n


\n<\/strong>2-<\/span> <\/strong>Kontrols\u00fcz \u00d6n ve Son Test Modeli:
\n<\/span><\/strong><\/span><\/p>\n

 <\/p>\n

 <\/p>\n

Bu modelde bir \u00f6nceki modelde kontrol edilmeyen grup elemanlar\u0131n\u0131n deney\/test \u00f6ncesi durumunun bilinmemesinin etkisi ortadan kald\u0131r\u0131lmaktad\u0131r. Ancak, \u00f6n test ve son test aras\u0131ndaki zamandan kaynakl\u0131 hatalar kontrol edilememektedir. Deney sonunda sonu\u00e7lar\u0131n anlaml\u0131 olup olmad\u0131\u011f\u0131na \u00f6ntest ve son test puanlar\u0131n\u0131n ortalamalar\u0131n\u0131n kar\u015f\u0131la\u015ft\u0131r\u0131ld\u0131\u011f\u0131 \u201cba\u011f\u0131ml\u0131 gruplar i\u00e7in t testi\u201d ile s\u0131nan\u0131r.<\/span><\/p>\n

 <\/p>\n

 <\/p>\n

\u00a0<\/strong><\/p>\n

 <\/p>\n

 <\/p>\n

B)<\/span> <\/strong>KONTROL GRUBU OLAN MODELLER
\n<\/span><\/strong><\/span><\/p>\n\n\n\n
\n
\n

Gd<\/sub> D T ( – ) Gk<\/sub> T
\n<\/span><\/span><\/strong><\/p>\n

 <\/p>\n<\/div>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n

 
\n1-<\/span> <\/strong>Kontroll\u00fc son test modeli:
\n<\/span><\/strong><\/span><\/p>\n

 <\/p>\n

Bu tip ara\u015ft\u0131rmada deneyi\/testi hatalardan ar\u0131nd\u0131rmak ve i\u00e7 ge\u00e7erlili\u011fi art\u0131rmak i\u00e7in deneye kontrol grubu sokulur. \u0131ki gruptan hangisinin deney ya da kontrol grubu olaca\u011f\u0131 rassal olarak belirlenir. Bu t\u00fcr modellerde deney grubu teste\/deneye tabi tutulur ve sonunda test edilir. Ayn\u0131 test, deney yap\u0131lmadan kontrol grubuna da uygulan\u0131r. \u0131ki grubun \u00f6l\u00e7\u00fcmleri aras\u0131ndaki fark\u0131n anlaml\u0131 olup olmad\u0131\u011f\u0131 \u201cba\u011f\u0131ms\u0131z gruplar i\u00e7in z y ada t testi\u201d ile s\u0131nan\u0131r.<\/span><\/p>\n

 <\/p>\n

 <\/p>\n

Dikkat: Deney ve kontrol gruplar\u0131 birbirine olabildi\u011fince \u00f6l\u00e7\u00fclmek istenen nitelik bak\u0131m\u0131ndan benzer olmas\u0131na dikkat edilmelidir. Yine gruplar\u0131n deneye ba\u015flamadan \u00f6nce aralar\u0131ndaki e\u015fitli\u011fin olup olmad\u0131\u011f\u0131 belli olmad\u0131\u011f\u0131 i\u00e7in bundan do\u011fan bir hata kayna\u011f\u0131 ihtimali vard\u0131r.<\/span><\/p>\n\n\n\n
\n
\n

Gd<\/sub> T\u00f6<\/sub> D Ts ( – ) Gk<\/sub> T\u00f6<\/sub> Ts
\n<\/span><\/span><\/strong><\/p>\n

 <\/p>\n<\/div>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n

2-<\/span> <\/strong>Kontroll\u00fc \u00f6n test ve son test modeli:
\n<\/span><\/strong><\/span><\/p>\n

 <\/p>\n

 <\/p>\n

Bu modelde deney ve kontrol gruplar\u0131 bir \u00f6n testte tabi tutulur ve \u00f6ntest puanlar\u0131n\u0131n olabildi\u011fince benzer, birbirine yak\u0131n olmas\u0131na dikkat edilir. Bir \u00f6nceki modeldeki hata kayna\u011f\u0131 elimine edilir b\u00f6ylece. Deney ve kontrol gruplar\u0131n\u0131n her biri i\u00e7in son test ile \u00f6ntest puanlar\u0131 aras\u0131ndaki fark hesaplan\u0131r ve bu farklar\u0131n ortalamas\u0131 al\u0131n\u0131r. Her iki grup i\u00e7in hesaplanan farklar\u0131n ortalamalar\u0131 aras\u0131nda anlaml\u0131 bir fark\u0131n olup olmad\u0131\u011f\u0131 \u201cba\u011f\u0131ms\u0131z gruplar i\u00e7in t testi\u201d ile s\u0131nan\u0131r.<\/span><\/p>\n

\u00a0<\/strong><\/p>\n

 <\/p>\n

 <\/p>\n

C)<\/span> <\/strong>\u00d6RNEKLEM VE KONTROL GRUPLU MODELLER
\n<\/span><\/strong><\/span><\/p>\n

 <\/p>\n

Bu t\u00fcr modellerde, \u00f6rnekleme teknikleri ile e\u015fitlenmi\u015f deney ve kontrol gruplar\u0131 vard\u0131r.<\/span><\/p>\n

 <\/p>\n\n\n\n
\n
\n

Gd<\/sub>R<\/sup> D T ( – ) Gk<\/sub>R<\/sup> T
\n<\/span><\/span><\/strong><\/p>\n

 <\/p>\n<\/div>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n

1-<\/span> <\/strong>\u00c7apraz Test Modeli:
\n<\/span><\/strong><\/span><\/p>\n

 <\/p>\n

 <\/p>\n

Deney ve kontrol gruplar\u0131 rassal \u00f6rnekleme teknikleri kullan\u0131larak e\u015fitlenmi\u015ftir. Deney ba\u015flamadan \u00f6nce kontrol grubuna test uygulan\u0131r, deney grubuna ise ayn\u0131 test deneyden sonra uygulan\u0131r. Ortalamalar\u0131 kar\u015f\u0131la\u015ft\u0131rmak i\u00e7in \u201cba\u011f\u0131ms\u0131z \u00f6rneklemler i\u00e7in t testi\u201d kullan\u0131l\u0131r.<\/span><\/p>\n

\u00a0<\/strong><\/p>\n

 <\/p>\n

 <\/p>\n

2-<\/span> <\/span><\/strong><\/p>\n\n\n\n
\n
\n

Gd<\/sub>R<\/sup> T\u00f6<\/sub> D Ts<\/sub> ( – ) Gk<\/sub>R<\/sup> T\u00f6<\/sub> Ts<\/sub>
\n<\/span><\/span><\/strong><\/p>\n

 <\/p>\n<\/div>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

\u00d6rneklem ve kontroll\u00fc, \u00d6n ve Son test Modeli:
\n<\/span><\/span><\/strong><\/p>\n

 <\/p>\n

Bu modelde, gruplar \u00f6rnekleme yoluyla se\u00e7ilmi\u015ftir. Se\u00e7ilen bu gruplardan rassal olarak deney ve kontrol gruplar\u0131 belirlenir. Test her iki gruba deney ba\u015flamadan \u00f6nce ayn\u0131 anda uygulan\u0131r. Deney grubunda deney bittikten sonra ayn\u0131 test her iki gruba ayn\u0131 anda uygulan\u0131r. B\u00f6ylece her iki grup da kontrol alt\u0131na al\u0131n\u0131r. Her iki grupta, zamandan kaynakl\u0131 \u00f6l\u00e7me hatalar\u0131ndan ayn\u0131 d\u00fczeyde etkileneceklerinden kar\u015f\u0131la\u015ft\u0131rmalarda sorun olmaz. \u0131\u00e7 ve d\u0131\u015f ge\u00e7erlilik sa\u011flanm\u0131\u015f olur. Bu modelde kullan\u0131lacak test <\/span><\/p>\n

 <\/p>\n

 <\/p>\n\n\n\n
\n
\n

Gd<\/sub>R<\/sup> D Ts<\/sub> ( – ) Gk<\/sub>R<\/sup> Ts<\/sub>
\n<\/span><\/span><\/strong><\/p>\n

 <\/p>\n<\/div>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n

3-<\/span> <\/strong>\u00d6rneklem ve kontroll\u00fc, Son test Modeli:
\n<\/span><\/strong><\/span><\/p>\n

 <\/p>\n

Bu modelde, \u00f6nceki modelde kullan\u0131lan \u00f6n testler kullan\u0131lmaz. Bu modelde kullan\u0131lacak test, son test modeli ile ayn\u0131d\u0131r. <\/span><\/p>\n

 <\/p>\n

 <\/p>\n

 <\/p>\n

 <\/p>\n

ARA\u015fTIRMA MODELL\u0131LER\u0131NDE KULLANILACAK \u0131STAT\u0131ST\u0131KLER\u0131 BEL\u0131RLEME \u00d6L\u00c7\u00dcTLER\u0131
\n<\/span><\/span><\/strong><\/p>\n

 <\/p>\n

 <\/p>\n

 <\/p>\n\n\n\n
\n
\n

Kullan\u0131lacak \u0131statistikleri Belirleme \u00d6l\u00e7\u00fctleri
\n<\/span><\/span><\/strong><\/p>\n<\/div>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n
\n
\u00b7 Tek de\u011fi\u015fkenli (X)
\n<\/span><\/strong>\u00b7 \u0131ki de\u011fi\u015fkenli ( X ve Y)
\n<\/span><\/strong>\u00b7 \u00c7ok de\u011fi\u015fkenli
\n<\/span><\/strong>(p>2, X, Y, Z\u2026.p)
\n<\/span><\/strong><\/div>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n
\n
De\u011fi\u015fken Say\u0131s\u0131
\n<\/span><\/span><\/strong><\/div>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n
\n
\u00b7 S\u0131n\u0131flama (kesikli)
\n<\/span><\/strong>\u00b7 S\u0131ralama (kesikli)
\n<\/span><\/strong>\u00b7 E\u015fit aral\u0131kl\u0131 (s\u00fcrekli)
\n<\/span><\/strong>\u00b7 Oranl\u0131 (s\u00fcrekli)
\n<\/span><\/strong><\/div>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n
\n
De\u011fi\u015fkenin \u00d6l\u00e7ek T\u00fcr\u00fc ya da Yap\u0131s\u0131
\n<\/span><\/span><\/strong><\/div>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n
\n
\u00b7 Tek \u00f6rneklem
\n<\/span><\/strong>\u00b7 \u0131ki \u00f6rneklem (Ba\u011f\u0131ml\u0131, Ba\u011f\u0131ms\u0131z)
\n<\/span><\/strong>\u00b7 K \u00f6rneklem
\n<\/span><\/strong>(Ba\u011f\u0131ml\u0131, Ba\u011f\u0131ms\u0131z)
\n<\/span><\/strong> <\/p>\n<\/div>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n
\n
Grup Say\u0131s\u0131
\n<\/span><\/span><\/strong><\/div>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n
\n
\u00b7 Etki
\n<\/span><\/strong>\u00b7 \u0131li\u015fki\/Benzerlik
\n<\/span><\/strong><\/div>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n
\n
\u0131statistiklerin \u0131\u015flevselli\u011fi
\n<\/span><\/span><\/strong><\/div>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n
\n
Parametrik mi Parametrik Olmayan m\u0131?
\n<\/span><\/span><\/strong><\/div>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n

 <\/p>\n

Bir ara\u015ft\u0131rma deseninde toplanan verilere hangi istatistiklerin uygulanabilir oldu\u011funu belirlemek i\u00e7in baz\u0131 \u00f6l\u00e7\u00fctler s\u00f6z konusudur. Uygun istatistiklerle ara\u015ft\u0131rmay\u0131 \u00e7\u00f6z\u00fcmlemek, ara\u015ft\u0131rman\u0131n g\u00fcvenirli\u011fini art\u0131rmakla birlikte sonu\u00e7lar\u0131n tutarl\u0131 bir \u015fekilde yorumlanmas\u0131n\u0131 da sa\u011flar. \u0131statistik kitaplar\u0131 incelendi\u011finde ya da ara\u015ft\u0131rmalara bak\u0131ld\u0131\u011f\u0131nda \u00e7e\u015fitli istatistiklerle kar\u015f\u0131la\u015fmak m\u00fcmk\u00fcnd\u00fcr. En \u00f6nemli soru: elimizdeki veri i\u00e7in hangi istatisti\u011fi kullanmal\u0131y\u0131z? Bunun i\u00e7in cevaplanmas\u0131 gereken birka\u00e7 soru vard\u0131r. Kullan\u0131labilecek istatistikler toplanan verinin parametrik ya da nonparametrik olmas\u0131na, de\u011fi\u015fken say\u0131s\u0131na, de\u011fi\u015fkenlerin \u00f6l\u00e7ek t\u00fcr\u00fcne, grup say\u0131s\u0131na ve problemde aranan sorunun i\u015flevselli\u011fine g\u00f6re farkl\u0131l\u0131k g\u00f6stermektedir. Bu ba\u011flamda istatistiksel analiz y\u00f6ntemlerinden k\u0131saca bahsedersedebiliriz.<\/span><\/p>\n

 <\/p>\n

 <\/p>\n

A)<\/span> <\/strong>TEK DE\u011e\u0130\u015eKENL\u0130 \u00d6L\u00c7\u00dcMLERE ANAL\u0130ZLER
\n<\/span><\/strong><\/span><\/p>\n

Tek de\u011fi\u015fkenli bir modelde hesaplanabilecek \u201cbetimsel\u201d istatistikler, \u00f6l\u00e7ek t\u00fcr\u00fcne g\u00f6re farkl\u0131l\u0131k g\u00f6stermektedir. A\u015fa\u011f\u0131daki tabloda \u00f6l\u00e7ek t\u00fcr\u00fcne g\u00f6re elde edilen verilerde yap\u0131labilecek istatistikler \u00f6zetlenmi\u015ftir.<\/span><\/p>\n

 <\/p>\n

TABLO1: Tek De\u011fi\u015fkenli \u00d6l\u00e7\u00fcmlerde kullan\u0131lan istatistikler (\u00d6l\u00e7ek t\u00fcr\u00fcne g\u00f6re)
\n<\/span><\/span><\/strong><\/p>\n\n\n\n\n\n\n\n
\u00d6l\u00e7ek T\u00fcr\u00fc
\n<\/strong><\/td>\n		Betimsel \u0131statistikler
\n<\/strong><\/td>\n<\/tr>\n
S\u0131n\u0131flama \u00f6l\u00e7e\u011fi
\n<\/span><\/td>\n		*Frekans ve
\n<\/span>*y\u00fczde da\u011f\u0131l\u0131mlar\u0131na ili\u015fkin s\u0131n\u0131flay\u0131c\u0131 tablolar,
\n<\/span>*S\u0131n\u0131flarda y\u0131\u011f\u0131l\u0131m\u0131n homojenli\u011fi test edilir,
\n<\/span>*Tepe de\u011fer (mod),
\n<\/span>*Deneysel olas\u0131l\u0131klara g\u00f6re her hangi bir da\u011f\u0131l\u0131ma uygunluk testi
\n<\/span>*Kategorilerin g\u00f6zlenme oranlar\u0131 belirlenerek oransal kar\u015f\u0131la\u015ft\u0131rmalar,
\n<\/span>*\u00c7apraz tablo bi\u00e7iminde g\u00f6sterilen de\u011fi\u015fkenler aras\u0131nda ba\u011f\u0131ms\u0131zl\u0131k analizi,
\n<\/span>*Parametrik olmayan<\/em> y\u00f6ntemlerle hipotezler test edilir.
\n<\/span> <\/td>\n<\/tr>\n
S\u0131ralama \u00f6l\u00e7ekli veriler
\n<\/span><\/td>\n		*Frekans ve
\n<\/span>*y\u00fczde da\u011f\u0131l\u0131mlar\u0131na ili\u015fkin s\u0131n\u0131flay\u0131c\u0131 tablolar,
\n<\/span>*S\u0131n\u0131flarda y\u0131\u011f\u0131l\u0131m\u0131n homojenli\u011fi test edilir,
\n<\/span>*Uygunluk testi,
\n<\/span>*Orta de\u011fer (ortanca, medyan)hesaplan\u0131r,
\n<\/span>*Dizili\u015flerin rasgeleli\u011fi test edilebilir,
\n<\/span>*Parametrik olmayan korelasyon analizi yap\u0131labilir,
\n<\/span>*S\u0131ralama \u00f6l\u00e7eklerinde uygulanan birliktelik katsay\u0131lar\u0131 hesaplanabilir.
\n<\/span>*Parametrik olmayan<\/em> y\u00f6ntemlerle hipotezler test edilir.
\n<\/span>
\n<\/span><\/td>\n<\/tr>\n
E\u015fit aral\u0131kl\u0131 veriler
\n<\/span><\/td>\n		*Frekans ve y\u00fczdeler,
\n<\/span>*Aritmetik ortalama, standart sapma, s. Hata gibi betimsel istatistikleri hesaplama,
\n<\/span>*Da\u011f\u0131l\u0131m varsay\u0131mlar\u0131 ile uygunluk testleri,
\n<\/span>*Regresyon-Korelasyon(neden sonu\u00e7 ili\u015fkileri i\u00e7in)
\n<\/span>*S\u0131ralama ve s\u0131n\u0131flama \u00f6l\u00e7eklerine indirgenerek bu \u00f6l\u00e7ekler \u00fczerinden hesaplanabilen t\u00fcm istatistikler ve testler yap\u0131labilir.
\n<\/span><\/td>\n<\/tr>\n
Oranl\u0131 \u00f6l\u00e7ekle
\n<\/span><\/td>\n		*T\u00fcm betimsel istatistikler,
\n<\/span>*Frekans tablolar\u0131,
\n<\/span>*Parametrik testlere uygun olan t\u00fcm testler,
\n<\/span>*Uygunluk testleri,
\n<\/span>*t testi,
\n<\/span>*ANOVA-ANCOVA
\n<\/span>*Neden-sonu\u00e7 ili\u015fkileri i\u00e7in Do\u011frusal\/Do\u011frusal olmayan basit regresyon\/korelasyon hesaplanabilir.
\n<\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n

 <\/p>\n

Parametrik Testler: <\/strong>\u0131lgili parametreye, belirli bir da\u011f\u0131l\u0131ma ve varyans kavram\u0131na dayanarak i\u015flemler yapan esnek olmayan<\/em> istatistiksel y\u00f6ntemlerdir. <\/span><\/span><\/p>\n

 <\/p>\n

Parametrik Testlerin Uygulanma Ko\u015fullar\u0131
\n<\/span><\/span><\/strong><\/p>\n

1-<\/span> De\u011fi\u015fken, pop\u00fcl\u00e2syonda normal ya da normale d\u00f6n\u00fc\u015ft\u00fcr\u00fclebilir da\u011f\u0131l\u0131m g\u00f6stermelidir<\/span><\/span><\/p>\n

2-<\/span> De\u011fi\u015fken aral\u0131kl\u0131 ya da oranl\u0131 \u00f6l\u00e7ekli olmal\u0131d\u0131r.<\/span><\/span><\/p>\n

3-<\/span> De\u011fi\u015fken parametreleri <\/span><\/span>m bilinmelidir<\/span><\/span><\/p>\n

4-<\/span> N birim say\u0131s\u0131 belirli bir de\u011ferden daha fazla olmal\u0131d\u0131r.<\/span><\/span><\/p>\n

 <\/p>\n

Parametrik verilerde istatistik\u00ee analizlerin yap\u0131labilmesi i\u00e7in verilerin normal da\u011f\u0131lmas\u0131 gerekmektedir. Normallik ko\u015fuluna ba\u011fl\u0131 olarak yap\u0131lan istatistiksel analizler \u015f\u00f6yle:<\/span><\/p>\n

 <\/p>\n

T testi<\/span><\/p>\n

Z testi<\/span><\/p>\n

Varyans Analizi<\/span><\/p>\n

Pearson korelasyon Analizi<\/span><\/p>\n

Regresyon analizinde ba\u011f\u0131ml\u0131 de\u011fi\u015fken<\/span><\/p>\n

Fakt\u00f6r analiz<\/span><\/p>\n

K\u00fcmeleme analizi<\/span><\/p>\n

Diskriminant analizi<\/span><\/p>\n

Yap\u0131sal e\u015fitlik modeli<\/span><\/p>\n

 <\/p>\n

\u00d6rneklem B\u00fcy\u00fckl\u00fcklerine g\u00f6re Normallik testleri ve Paket Program\u0131
\n<\/span><\/span><\/strong><\/p>\n\n\n\n\n\n\n\n\n
\u00d6rneklem B\u00fcy\u00fckl\u00fc\u011f\u00fc
\n<\/strong><\/td>\n		Normallik Testi
\n<\/strong><\/td>\n		Program
\n<\/strong><\/td>\n<\/tr>\n
n<50<\/td>\nShapiro-Wilk testi<\/td>\nSPSS<\/td>\n<\/tr>\n
n>50<\/td>\nLiliefors testi<\/td>\nSPSS<\/td>\n<\/tr>\n
Her B\u00fcy\u00fckl\u00fck i\u00e7in<\/td>\nKolmogrov-Smirnov Testi<\/td>\nSPSS<\/td>\n<\/tr>\n
n<2000<\/td>\nShapiro-Wilk testi<\/td>\nSAS<\/td>\n<\/tr>\n
n>2000<\/td>\nKolmogrov-Smirnov Testi<\/td>\nSAS<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n

Parametrik Olmayan Testler: <\/strong>Parametreye, belirli bir da\u011f\u0131l\u0131ma ve varyansa dayanmadan i\u015flemler yapan genellikle veriler yerine onlar\u0131n s\u0131ralama puanlar\u0131n\u0131 kullanarak i\u015flem yapan esnek istatistiksel y\u00f6ntemlerdir.<\/span><\/span><\/p>\n

 <\/p>\n

Parametrik Olmayan Testlerin Uygulanma Ko\u015fullar\u0131
\n<\/span><\/span><\/strong><\/p>\n

2-<\/span> De\u011fi\u015fkenin herhangi bir da\u011f\u0131l\u0131ma uygun de\u011fildir.<\/span><\/span><\/p>\n

3-<\/span> De\u011fi\u015fken s\u0131n\u0131flama ya da s\u0131ral\u0131 \u00f6l\u00e7ekli olmal\u0131d\u0131r.<\/span><\/span><\/p>\n

4-<\/span> De\u011fi\u015fken parametreleri <\/span><\/span>m bilinmez.<\/span><\/span><\/p>\n

5-<\/span> G\u00f6zlemlerin homojen yap\u0131 olu\u015fturmaz.<\/span><\/span><\/p>\n

6-<\/span> N birim say\u0131s\u0131 belirli bir de\u011ferde olmas\u0131 gerekmez.<\/span><\/span><\/p>\n

 <\/p>\n

Tek De\u011fi\u015fkenli Parametrik Testler
\n<\/span><\/span><\/strong><\/p>\n

 <\/p>\n

 <\/p>\n

o<\/span> Tek \u00f6rneklem testleri: <\/strong>z ve t testi
\n<\/strong><\/span><\/span>o<\/span> \u0131ki \u00f6rneklem testleri:
\n<\/span><\/span><\/strong>o<\/span> Ba\u011f\u0131ms\u0131z \u00f6rneklemler. <\/strong>Z ve t testi
\n<\/strong><\/span><\/span>o<\/span> Ba\u011f\u0131ml\u0131 \u00f6rneklemler: <\/strong>E\u015fle\u015ftirilmi\u015f t testi
\n<\/strong><\/span><\/span>o<\/span> K-\u00d6rneklem Testleri:
\n<\/span><\/span><\/strong>o<\/span> Ba\u011f\u0131ms\u0131z k \u00f6rneklem: <\/strong>Tek y\u00f6nl\u00fc ANOVA, GLM
\n<\/strong><\/span><\/span>o<\/span> Ba\u011f\u0131ml\u0131 k \u00f6rneklem: <\/strong>\u0131ki y\u00f6nl\u00fc ANOVA, GLM
\n<\/strong><\/span><\/span>\u00d8<\/span> Do\u011frusal Ba\u011f\u0131nt\u0131 ve \u0131li\u015fki Analizi
\n<\/span><\/span><\/strong><\/p>\n

\u00b7<\/span> Basit Do\u011frusal regresyon ve korelasyon<\/span><\/p>\n

\u00b7<\/span> \u00c7oklu regresyon ve korelasyon<\/span><\/p>\n

\u00b7<\/span> Do\u011frusalla\u015ft\u0131r\u0131labilen regresyon y\u00f6ntemleri<\/span><\/p>\n

\u00d8<\/span> Do\u011frusal Olmayan Ba\u011f\u0131nt\u0131 ve \u0131li\u015fki Analizi
\n<\/span><\/span><\/strong><\/p>\n

\u00b7<\/span> Nonlinear, Lojistik, Probit, Geometrik, Polinomial Regresyon<\/span><\/p>\n

 <\/p>\n

 <\/p>\n

 <\/p>\n

 <\/p>\n

Tek De\u011fi\u015fkenli Parametrik Olmayan Testler
\n<\/span><\/span><\/strong><\/p>\n

 <\/p>\n

 <\/p>\n

o<\/span> Tek \u00f6rneklem testleri: <\/strong>Diziler testi, Binomial test, i\u015faret testi
\n<\/strong><\/span><\/span>o<\/span> \u0131ki \u00f6rneklem testleri:
\n<\/span><\/span><\/strong>o<\/span> Ba\u011f\u0131ms\u0131z \u00f6rneklemler: <\/strong>Mann-Whitney U testi
\n<\/strong><\/span><\/span>o<\/span> Ba\u011f\u0131ml\u0131 \u00f6rneklemler: <\/strong>Wilcoxon T testi
\n<\/strong><\/span><\/span>o<\/span> K-\u00d6rneklem Testleri:
\n<\/span><\/span><\/strong><\/p>\n

o<\/span> Ba\u011f\u0131ms\u0131z k \u00f6rneklem: <\/strong>Kruskal-Wallis<\/span><\/span><\/p>\n

o<\/span> Ba\u011f\u0131ml\u0131 k \u00f6rneklem: <\/strong>Fredman \u0131ki Y\u00f6nl\u00fc Varyans Analizi, Medyan Testi<\/span><\/span><\/p>\n

 <\/p>\n

 <\/p>\n

\u00d8<\/span> Ba\u011f\u0131nt\u0131, \u0131li\u015fki ve Uyumluk Analizi: <\/strong>Spearman, Kendal, Taub<\/sub>, Cohen Kappa, Kendal Tauc<\/sub><\/span><\/span><\/p>\n

 <\/p>\n

 <\/p>\n

 <\/p>\n

 <\/p>\n

B)<\/span> <\/strong>\u0131K\u0131 DE\u011f\u0131\u015fKENL\u0131 \u00d6L\u00c7\u00dcMLERE ANAL\u0131ZLER
\n<\/span><\/strong><\/span><\/p>\n

\u0131ki De\u011fi\u015fkenli \u00f6l\u00e7\u00fcmler \u00fczerinde iki de\u011fi\u015fken aras\u0131nda (X ve Y) (1) ili\u015fkinin d\u00fczeyine ve (2) fark\u0131n miktar\u0131 test edilebilir. A\u015fa\u011f\u0131daki tabloda 2 de\u011fi\u015fkenli \u00f6l\u00e7\u00fcmlerin \u00f6l\u00e7ek t\u00fcr\u00fcne g\u00f6re yap\u0131labilecek istatistikler \u00f6zetlenmi\u015ftir.<\/span><\/p>\n

TABLO 2: De\u011fi\u015fken Say\u0131s\u0131 \u0131ki ise hesaplanacak olan \u0131li\u015fki Analizleri (korelasyon)
\n<\/span><\/span><\/strong><\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
1.De\u011fi\u015fken
\n<\/span><\/span><\/strong><\/td>\n		2.De\u011fi\u015fken
\n<\/span><\/span><\/strong><\/td>\n		\u0131statistik
\n<\/span><\/span><\/strong><\/td>\n		A\u00e7\u0131klama
\n<\/span><\/span><\/strong><\/td>\n<\/tr>\n
S\u0131n\u0131flama (ikili)
\n<\/span><\/span><\/em><\/p>\n
\n

 <\/p>\n<\/div>\n

\n

 <\/p>\n<\/div>\n<\/td>\n

S\u0131n\u0131flama (ikili)
\n<\/span><\/span><\/em><\/td>\n		Phi Katsay\u0131s\u0131
\n<\/span><\/span><\/em><\/td>\n		Ger\u00e7ek \u0131kili
\n<\/span><\/span><\/em><\/td>\n<\/tr>\n
S\u0131n\u0131flama (ikili)
\n<\/span><\/span><\/em><\/td>\n		Tetrakorik
\n<\/span><\/span><\/em><\/td>\n		Yapay \u0131kili
\n<\/span><\/span><\/em><\/td>\n<\/tr>\n
E\u015fit Aral\u0131kl\u0131\/Oranl\u0131
\n<\/span><\/span><\/em><\/td>\n		Nokta \u00c7ift serili
\n<\/span><\/span><\/em><\/td>\n		Ger\u00e7ek ikili
\n<\/span><\/span><\/em><\/td>\n<\/tr>\n
S\u0131n\u0131flama (ikili\/\u00e7oklu
\n<\/span><\/span><\/em><\/td>\n		Cramer V
\n<\/span><\/span><\/em><\/td>\n		<\/td>\n<\/tr>\n
S\u0131n\u0131flama (\u00e7oklu)
\n<\/span><\/span><\/em><\/td>\n		S\u0131n\u0131flama (ikili)
\n<\/span><\/span><\/em><\/td>\n		Kontenjan Kats.
\n<\/span><\/span><\/em><\/td>\n		<\/td>\n<\/tr>\n
S\u0131n\u0131flama
\n<\/span><\/span><\/em><\/td>\n		S\u0131n\u0131flama
\n<\/span><\/span><\/em><\/td>\n		Lamda
\n<\/span><\/span><\/em><\/td>\n		Ba\u011f\u0131ml\u0131\/Ba\u011f\u0131ms\u0131z
\n<\/span><\/span><\/em><\/td>\n<\/tr>\n
S\u0131n\u0131flama (ikili\/\u00e7oklu)
\n<\/span><\/span><\/em><\/td>\n		E\u015fit Aral\u0131kl\u0131\/Oranl\u0131
\n<\/span><\/span><\/em><\/td>\n		\u00c7ift serili
\n<\/span><\/span><\/em>Eta
\n<\/span><\/span><\/em><\/td>\n		Yapay ikili
\n<\/span><\/span><\/em>E\u011frisel ili\u015fkiler
\n<\/span><\/span><\/em><\/td>\n<\/tr>\n
S\u0131ral\u0131
\n<\/span><\/span><\/em><\/p>\n
\n

 <\/p>\n<\/div>\n

\n

 <\/p>\n<\/div>\n<\/td>\n

S\u0131ral\u0131
\n<\/span><\/span><\/em><\/p>\n
\n

 <\/p>\n<\/div>\n

\n

 <\/p>\n<\/div>\n<\/td>\n

Sperman rho
\n<\/span><\/span><\/em><\/td>\n		Yapay ikili
\n<\/span><\/span><\/em><\/td>\n<\/tr>\n
Kendall tau
\n<\/span><\/span><\/em><\/td>\n		N<10
\n<\/span><\/span><\/em><\/td>\n<\/tr>\n
Goodman
\n<\/span><\/span><\/em><\/td>\n		<\/td>\n<\/tr>\n
Somer D
\n<\/span><\/span><\/em><\/td>\n		<\/td>\n<\/tr>\n
E\u015fit Aral\u0131kl\u0131\/Oranl\u0131
\n<\/span><\/span><\/em><\/td>\n		S\u0131ral\u0131
\n<\/span><\/span><\/em><\/td>\n		Sperman rho
\n<\/span><\/span><\/em><\/td>\n		Madde Analizi
\n<\/span><\/span><\/em><\/td>\n<\/tr>\n
E\u015fit Aral\u0131kl\u0131\/Oranl\u0131
\n<\/span><\/span><\/em><\/td>\n		Pearson r
\n<\/span><\/span><\/em><\/td>\n		S\u00fcrekli de\u011fi\u015fkenler
\n<\/span><\/span><\/em><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n

 <\/p>\n

\u0131\u015fLEVLER\u0131NE G\u00d6RE \u0131STAT\u0131ST\u0131KSEL TESTLER<\/span><\/strong><\/p>\n

 <\/p>\n

Bir ba\u015fka s\u0131n\u0131flama ise i\u015flevlerine g\u00f6redir kullan\u0131labilecek istatistikler olarak a\u015fa\u011f\u0131daki tabloda \u00f6zetlenmi\u015ftir. <\/span><\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n
\u0131\u015flevlerine G\u00f6re \u0131statistiksel Testler
\n<\/span><\/span><\/strong><\/td>\n<\/tr>\n
1) Ortalama Fark\/ Farklar\u0131n \u00d6nemlili\u011fi<\/span><\/td>\nZ testi, Student t testi, ANOVA, Hortelling-T2<\/sup> testi, MANOVA<\/span><\/td>\n<\/tr>\n
2) Oranlararas\u0131 Fark\/Farklar\u0131n \u00d6nemlili\u011fi <\/span><\/td>\nZ testi, Student t testi, kikare testi<\/span><\/td>\n<\/tr>\n
3) \u0131ki ve \u00e7oklu kategoriler aras\u0131 k\u00fcmelenmenin\/y\u0131\u011f\u0131lman\u0131n \u00d6nemlili\u011fi<\/span><\/td>\nX2<\/sup> Testi, Loglinear Analiz, Uyum Analizi, Homojenite Analizi<\/span><\/td>\n<\/tr>\n
4) Meydanlar\u0131n benzerli\u011fi <\/span><\/td>\nMann Whitney U testi, Kruskal Wallis H testi, Cochran Testi, Moses Extreme Reactions test<\/span><\/td>\n<\/tr>\n
5) Da\u011f\u0131l\u0131mlar\u0131n benzerli\u011fi testi<\/span><\/td>\nX2<\/sup> testi, Kolmogrov Smirnov testi, Shapiro-Wilks testi,Anderson-Darling testi<\/span><\/td>\n<\/tr>\n
6) \u0131li\u015fki d\u00fczeyinin \u00f6nemlili\u011fi<\/span><\/td>\nPearson, Spearman, Kendal, Setler aras\u0131, Parametrik olmayan setler aras\u0131 korelasyon analizleri<\/span><\/td>\n<\/tr>\n
7) Ba\u011f\u0131nt\u0131n\u0131n \u00f6nemlili\u011fi <\/span><\/td>\n Basit regresyon, \u00c7oklu regresyon, Lojistik, E\u011fim Analizi, Cox Regresyon Analizi<\/span><\/span><\/td>\n<\/tr>\n
8) Uyumlulu\u011fun-Uyu\u015fumun \u00d6nemlili\u011fi<\/span><\/td>\nCohen Kappa, Kendal Concordance analizi, Uyum analizi<\/span><\/td>\n<\/tr>\n
9) Prototip Belirleme\/S\u0131n\u0131fland\u0131rma ve Atama Analizleri i\u00e7in<\/span><\/td>\n K\u00fcmeleme, Ay\u0131rma, \u00c7ok boyutlu \u00f6l\u00e7ekleme analizi<\/span><\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n

 <\/p>\n

Unutulmamal\u0131d\u0131r ki, ara\u015ft\u0131rman\u0131n problem c\u00fcmlesini \u00e7\u00f6zebilecek birden fazla istatistik olabilir. Burada \u00f6nemli olan problem c\u00fcmlesine uygun ve toplanan verilerin niteli\u011fine uygun istatistikleri kullanabilmektir.<\/span><\/p>\n

 <\/p>\n

KAYNAK\u00c7A
\n<\/span><\/span><\/span>\u00d6zdamar, K. (2002). Paket Programlar ile \u0131statistiksel Veri Analizi 2<\/strong>. <\/em>Eski\u015fehir: ETAM A.\u015f. Matbaa
\n<\/span>Tesisleri, Kaan Kitapevi
\n<\/span><\/p>\n

 <\/p>\n

G\u00f6nderen:
\n<\/em><\/p>\n

T\u00fclin Otbi\u00e7er ACAR
\ntulinotbicer@yahoo.com Hacettepe \u00dcniversitesi Doktora \u00d6\u011frencisi
\nAhmet Yesevi \u00dcniversitesi, \u00d6l\u00e7me ve De\u011ferlendirme Uzman\u0131”<\/em><\/p>\n","protected":false},"excerpt":{"rendered":"

ARA\u015eTIRMA MODELLER\u0130 VE KULLANILAN \u0130STAT\u0130ST\u0130KLER       ARA\u015fTIRMA MODELLER\u0131 A) Kontrol Grubu ve \u00d6rneklemi Olmayan Modeller B) Kontrol Grubu Olan Modeller C) \u00d6rneklem Ve Kontrol Gruplu Modeller 1- Kontrols\u00fcz Son Test Modeli: 1- Kontroll\u00fc son test modeli: 1- \u00c7apraz Test Modeli: 2-Kontrols\u00fcz \u00d6n ve Son Test Modeli: 2- Kontroll\u00fc \u00f6n test ve son test…<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_kad_post_transparent":"","_kad_post_title":"","_kad_post_layout":"","_kad_post_sidebar_id":"","_kad_post_content_style":"","_kad_post_vertical_padding":"","_kad_post_feature":"","_kad_post_feature_position":"","_kad_post_header":false,"_kad_post_footer":false,"_kad_post_classname":"","footnotes":""},"categories":[],"tags":[],"class_list":["post-100","post","type-post","status-publish","format-standard","hentry"],"_links":{"self":[{"href":"https:\/\/www.istatistik.gen.tr\/index.php?rest_route=\/wp\/v2\/posts\/100","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.istatistik.gen.tr\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.istatistik.gen.tr\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.istatistik.gen.tr\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.istatistik.gen.tr\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=100"}],"version-history":[{"count":1,"href":"https:\/\/www.istatistik.gen.tr\/index.php?rest_route=\/wp\/v2\/posts\/100\/revisions"}],"predecessor-version":[{"id":112,"href":"https:\/\/www.istatistik.gen.tr\/index.php?rest_route=\/wp\/v2\/posts\/100\/revisions\/112"}],"wp:attachment":[{"href":"https:\/\/www.istatistik.gen.tr\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=100"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.istatistik.gen.tr\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=100"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.istatistik.gen.tr\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=100"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}